High-resolution Range Profiles Recognition Based on SAR/ISAR Images

摘要

High-resolution radar range profile (HRRP) recognition has attracted a great attention for its real-time ability in imaging and processing. However, HRRP is very sensitive to aspect angle. Traditional methods, using range profile itself to create template set, require a large amount of targets' range profiles in as many as aspect angles. For non-cooperated target, these methods can't offer satisfactory recognition results. To amend the problem, in this paper, we propose a HRRP recognition method based on SAR/ISAR images. The method gets inspiration from the facts that SAR/ISAR images contain cross-range information of targets and it's less sensitive to aspect angle and it's able to detect target at large distances. This method collects the SAR/ISAR images of targets as template set. By extracting the scattering centers and projecting to the line of sight, the HRRP template set can be obtained. Results on computational electromagnetics show that this method has a better robustness than the method using range profiles as templates when the line-of-sight has deviation or the frequency band is different between the template and the tested range profiles.